CN101114988B - Flow Control Algorithm Based on Discontinuous Transmission Prediction for Adaptive Multi-Rate Service - Google Patents

Abstract

The invention discloses a flow control algorithm based on a discontinuous emission adaptive multi-rate prediction service, which is invented to solve the problem of resource waste of the voice service bandwidth in the prior art, comprising the following procedures: first, the discontinuous emission statuses of the voice service in a prediction window are counted, then the average status of the discontinuous emission statuses in the prediction window are counted; then the average rate of the discontinuous emission statuses in the prediction window is counted; the bandwidth prediction values of all the voice services are added and the total bandwidth of the voice service distribution of the voice service is obtained in a next dispatching cycle. If the total bandwidth of the voice service distribution is smaller than the total transmission bandwidth, then the bandwidth of the voice service that can be used in the next dispatching cycle is the total bandwidth of the voice service distribution, and the surplus transmission bandwidth is distributed to other types of services. The invention enhances the resource utilization ratio and system capacity of the transmission bandwidth of the system.

Description

Flow Control Algorithm Based on Discontinuous Transmission Prediction for Adaptive Multi-Rate Service

technical field

The invention relates to a flow control strategy of a Wideband Code Division Multiple Access (WCDMA) system (hereinafter referred to as WCDMA) RNC (Radio Network Controller) network element to improve bandwidth utilization and system performance.

Background technique

In the WCDMA system, the business types are divided into conversational business, interactive business, stream business and background business. These four types have different priorities. In actual system application, resources are limited. As shown in the schematic diagram of the system in Figure 1, for downlink services, the communication strategy of renting E1 lines is usually adopted between the network element RNC and the network element NodeB. The number of leased E1 lines is limited. Therefore, when the system carries mixed services (4 types of services), it is necessary to divide the limited transmission resources according to the nature of the current various types of traffic and the requirements for QoS. , in order to ensure the service performance that requires high communication quality. For example, the bandwidth for conversational services is allocated first, then the bandwidth for stream services is allocated, and finally the bandwidth for interactive and background services is allocated.

For conversational services, they can be divided into two types: fixed bit rate and variable bit rate (AMR). When conversational services adopt AMR coding, the service rate is related to the system load and transmission power. It can be determined that when the average rate is relatively low within a certain period, it means that the system load is relatively high or the transmission power is relatively large. For voice services, a relatively low coding rate will be adopted, so it can be predicted that a relatively low coding rate will also be adopted in the next TTI.

According to the research results of international organizations such as IEEE, for conversational services, there are about 50% silent periods during the conversation, during which no data is transmitted during the conversation, and the probability is called the voice activation factor Alfa. During the conversation, according to the activation factor Alfa, it migrates between two states, the active state and the inactive state. The state transition obeys the Markov random process, and the state transition is shown in Figure 2. When the voice is activated, the rate of the voice service is determined by the AMR code.

带宽资源浪费。Assuming that the average bit rate of the AMR voice service in a certain statistical period is BitRate _i , and the total amount of voice services in the system is N, if the discontinuous transmission distribution characteristics of the voice are not considered, in order to first ensure the quality of service communication, it must be in In the total transmission bandwidth resource W, ∑BitRate _i is assigned to give priority to the voice service. If the voice activation factor is Alfa, then, from an average point of view, it will cause

Waste of bandwidth resources.

Contents of the invention

Aiming at the defects and deficiencies of the prior art, the purpose of the present invention is to provide a flow control algorithm based on discontinuous transmission prediction adaptive multi-rate services that can improve the resource utilization rate and system capacity of WCDMA system transmission bandwidth.

In order to achieve the above object, the present invention adopts the following technical solutions: a flow control algorithm based on discontinuous transmission prediction adaptive multi-rate services, comprising the following steps:

Step A, setting the prediction window and the discontinuous emission comparison threshold;

Step B, counting the discontinuous transmission status of the voice service within the prediction window;

Step C, according to the statistical record, calculate the average state of the discontinuous transmission state in the prediction window; compare the average state of the discontinuous transmission state with the discontinuous transmission comparison threshold, and judge whether the voice service is in the next scheduling period There is data to be sent, if the average state of the discontinuous transmission state is less than the discontinuous transmission comparison threshold, no bandwidth will be allocated for the voice service in the next scheduling period, if the average state of the discontinuous transmission state is greater than or equal to the If the discontinuous transmission comparison threshold is used, the voice service will be allocated bandwidth in the next scheduling cycle;

Step D, calculating the bandwidth within the prediction window;

Step E, adding up the bandwidths of all voice services to obtain the voice service allocation total bandwidth of the voice services in the next scheduling period;

Step F, if the total allocated bandwidth of the voice service is greater than the total transmission bandwidth, then the bandwidth that can be used by the voice service in the next scheduling cycle is the total transmission bandwidth; if the total bandwidth allocated to the voice service is less than the total transmission bandwidth, then the next The bandwidth that can be used by the voice service in one scheduling period allocates the total bandwidth for the voice service, and then allocates the remaining transmission bandwidth to other types of services.

The method of the invention can effectively solve the voice service flow control strategy in the case of limited transmission bandwidth resources in the WCDMA system, and improve the resource utilization rate of the system transmission bandwidth and the system capacity.

Description of drawings

Figure 1 Schematic diagram of wireless communication system

Figure 2 Schematic diagram of the state transition model

Figure 3Amr Voice Service Resource Prediction Flowchart

Detailed ways

Aiming at the waste of voice service bandwidth resources pointed out in the background technology, the present invention adopts a flow control strategy that predicts the total amount of bandwidth required by users of multiple AMR services in the next scheduling period, and uses the predicted value as the amount of resource application to improve system transmission. The utilization rate of the bandwidth increases the capacity of the system.

Below in conjunction with accompanying drawing 3 the present invention is described in further detail:

1) For service i, in the scheduling period k, record the discontinuous transmission state DtxState[k] in the current scheduling period, if there is data transmission for service i in the current period, then DtxState[k]=1, otherwise it is 0

2) Before the arrival of the next scheduling period, all services send a flow control request frame to the flow control scheduling center, and the information carried in the frame is a discontinuous transmission state sequence DtxState[k], where k=1, 2, 3, 4..., T, where T is the statistical period, set to 15

3) The flow control center calculates the average discontinuous transmission state in the window T of each service i according to the discontinuous transmission information in the flow control request frame sent by each service i

4) If MeanDtxState _i ＞=DtxThresHold, then the flow control center allocates bandwidth resource BitRate for service i, otherwise it is 0

5) Calculate the bandwidth resource BitRate, the calculation process is

Calculate the total bandwidth usage within 15 TTIs of the statistics period

Calculate the number of speech frames within 15 TTIs

Computing Bandwidth Resources

6) The flow control center adds up the bandwidth allocated by all services to obtain a total voice service allocated bandwidth W _C , and before the next scheduling cycle, the flow control center uses the flow control allocation frame to send W _C to all services

7) For each service i, in the next scheduling cycle, if there is data to be sent, as long as W _C > 0, then the corresponding voice frame is sent, and W _C = W _C -MaxBitRate, until W _C is 0 (MaxBitRate sum of predicted bandwidths for all AMR traffic)

To sum up, the present invention can effectively utilize the limited transmission bandwidth resource in the WCDMA system by adopting the flow control method based on the discontinuous transmission of the predicted voice service, and improves the system capacity.

Claims (1)

1. the flow control algorithm based on discontinuous emission prediction adaptive multi-rate service comprises the steps:

Steps A is provided with prediction window and discontinuous emission comparison threshold;

Step B, the discontinuous emission state of statistics speech business in described prediction window;

Step C according to the statistics record, calculates the mean state of the discontinuous emission state in the described prediction window; The mean state and the described discontinuous emission comparison threshold of this discontinuous emission state are compared, judge whether this speech business has data to need to send in next dispatching cycle, if the mean state of this discontinuous emission state is less than described discontinuous emission comparison threshold, then in next dispatching cycle, do not distribute bandwidth for this speech business, if the mean state of this discontinuous emission state more than or equal to described discontinuous emission comparison threshold, is then distributed bandwidth for this speech business in next dispatching cycle;

Step D, computation bandwidth in described prediction window;

Step e with the bandwidth addition of all speech businesses, obtains the speech business of all speech businesses in next dispatching cycle and distributes total bandwidth;

Step F, if described speech business distributes total bandwidth greater than the transmission bandwidth total amount, the bandwidth that so next speech business dispatching cycle can be used is the transmission bandwidth total amount; If described speech business distributes total bandwidth less than the transmission bandwidth total amount, the bandwidth that so next speech business dispatching cycle can be used is distributed total bandwidth as speech business, then remaining transmission bandwidth is distributed to the business of other type.

Images